Why do different workstations keep losing their domain trust?

November 17, 2026 5 answers
Asked by u.592+signalnotch

We have about 500 Windows devices on an Active Directory domain, and a few different machines report a trust relationship failure each week. Removing and rejoining them fixes it, but secure channel repair commands haven't worked for us. What should we check to find the cause?

Answers

_y63!bytepulley.8

I'd stop treating each rejoin as the end of the job and collect evidence from a couple of affected machines before changing them. Check the client and domain controller event logs around the failure, and enable verbose Netlogon logging if the existing events aren't enough. Note which domain controller each client is contacting.

Then check domain controller health and replication, especially if there are multiple controllers. An isolated or outdated controller, or a restored controller accidentally connected to the production network, is worth ruling out. A rejoin can get a workstation working without explaining why its machine account secret got out of sync.

Reply to this answer
_a96+voltspool.3

Check for duplicate computer names or an imaging process that leaves multiple devices using the same computer account. Two machines competing for one account can knock each other out of sync. The fact that the affected devices vary doesn't rule that out, though it's only one possibility.

Reply to this answer
2m~relaycaddy_493

For a repair attempt, sign in with a local admin account and try Test-ComputerSecureChannel -Repair -Credential DOMAIN\Admin -Verbose from an elevated PowerShell session. Reset-ComputerMachinePassword is another option, and it can target a particular domain controller.

Since repair has already failed in your environment, capture the verbose output and the relevant events rather than just running it again. The error may help distinguish a connectivity or authentication problem from a secure channel that needs resetting. Rejoining is still a fallback when repair won't work.

Reply to this answer
5x+nodewrench_813

Look at the basics too: DNS resolution and domain controller discovery, network connectivity, and time sync on both clients and controllers. DHCP lease and DNS scavenging settings are worth reviewing if records are going missing or pointing at the wrong devices.

Those checks can uncover problems that interfere with contacting a controller, but a trust error alone doesn't prove DNS or time is the cause.

Reply to this answer
8p!foldcaddy_617

A reboot may be a quick thing to try on an affected workstation before removing it from the domain. Also compare failure dates with Windows updates and deployment changes, but don't assume an update is responsible without a pattern in the logs or device versions.

Reply to this answer

Add your answer

Answers are reviewed before publication.

Related Discussions